Abstract
Nimorazole belongs to the imidazole-based family of antibiotics to fight against anaerobic bacteria. Moreover, nimorazole is now in Phase 3 clinical trial in Europe for potential use as a hypoxia radiosensitizer for treatment of head and neck cancers. We envision the use of [15N3]nimorazole as a theragnostic hypoxia contrast agent that can be potentially deployed in the next-generation MRI-LINAC systems. Here, we report the first but very important steps to create long-lasting (for tens of minutes) hyperpolarized state on three 15N sites of [15N3]nimorazole with T1 of up to ~6 minutes. The nuclear spin polarization was boosted by ~67000-fold at 1.4 T (corresponding to P15N of 3.2%) via 15N-15N spin-relayed SABRE-SHEATH hyperpolarization technique relying on simultaneous exchange of [15N3]nimorazole and parahydrogen on polarization transfer Ir-IMes catalyst. The isomeric side product of [15N3]nimorazole synthesis was hyperpolarized too. The presented results pave the way to efficient spin-relayed SABRE-SHEATH hyperpolarization of a wide range of imidazole-based antibiotics and chemotherapeutics.
Keywords: parahydrogen, hyperpolarization, SABRE-SHEATH, nimorazole, radiosensitization
Graphical Abstract

Synthesis of 15N3-enriched antibiotic and hypoxic radiosensitizer nimorazole is reported. Spin-relayed Signal Amplification By Reversible Exchange in microtesla magnetic field yielded up to 3.2% 15N nuclear spin polarization with relaxation time constant of 5.9 ± 0.2 min at 1.4 T. The isomer of nimorazole formed in the synthesis was also hyperpolarized indicating that this approach can be extended to a wide range of nitroimidazole-based drugs.
Introduction
Equilibrium nuclear spin polarization (P) of key biologically-relevant nuclei such as 1H, 13C and 15N is on the order of 10−6-10−5 at clinically relevant conditions of body temperature and the magnetic field of up to 3 T. Since signal-to-noise ratio (SNR) is directly proportional to P in magnetic resonance (MR), NMR techniques including most notably MRI have inherently low detection sensitivity. As a result, clinical MRI is typically limited to imaging of protons of water and lipids, which have high physiological concentrations of tens of moles/L. In contrast, MR imaging of dilute metabolites is a grand challenge despite potential benefits of providing rich metabolic information, which is central to diagnosis of diseases with aberrant metabolism such as cancer, diabetes, etc. NMR hyperpolarization techniques allow transiently increasing P up to the order of unity.[1, 2] This massive increase in sensitivity gains by 4–5 orders of magnitude can be employed for metabolic imaging of hyperpolarized (HP) compounds, which are typically injected and act as metabolic contrast agents.[3–5] 13C-HP [1-13C]pyruvate is the most notable metabolite that has been developed so far to probe upregulated glycolysis in cancer and many other diseases.[6–11] Numerous clinical trials are now enabled because access to this HP contrast agent became possible via dissolution Dynamic Nuclear Polarization (d-DNP).[12–14] Despite many technology breakthroughs[15–17] over the recent years, d-DNP has many shortcomings of high equipment- (>$2M) and operational costs, and long polarization times.[12] For example, hyperpolarization of 15N sites has been a main challenge for conventional d-DNP requiring custom 1H→15N cross-polarization hardware to achieve efficient hyperpolarization with P15N>15%.[18] A number of alternative hyperpolarization techniques are now under development to address d-DNP limitations.[1, 19]
Signal Amplification by Reversible Exchange (SABRE)[20–22] technique relies on the simultaneous chemical exchange of parahydrogen (p-H2) and to-be-hyperpolarized biomolecule on a hexacoordinate Ir complex, usually possessing IMes or a similar ligand.[23] At the appropriate magnetic field, nuclear spin order of parahydrogen-derived protons is transferred to the nuclei of coordinated substrate. Subsequent dissociation of the complex liberates HP biomolecule to the solution. SABRE technique underwent a rapid development over the past few years,[19, 24] and the list of amenable biomolecules has been rapidly expanding.[25–30] Although [1-13C]pyruvate and other 13C-containing biomolecules are biomedically interesting, their primary limitation is a relatively short lifetime of HP state with exponential decay time constant T1 on the order of 1 min in most favorable cases.[4, 5, 12] HP 15N sites have garnered a lot of attention recently because their T1 can be on the order of 10 min or longer [28, 31–33] – as a result, the detection window of HP injectable biomolecules can be significantly expanded.[34, 35]
Although there have been some reports for hyperpolarization of 15N-labeled biomolecules via d-DNP, this technology is not well suited due to ultra-long polarization times of several hours, and low polarization yields on top of the other d-DNP limitations.[35, 36] As a result, the progress of biomedical translation of HP 15N has been relatively slow.
A variant of SABRE technique called SABRE-SHEATH (SABRE in SHield Enables Alignment Transfer to Heteronuclei, Figure 1)[37, 38] allows for very efficient polarization of 15N sites in 15N-containing compounds with fast polarization times (1 min or less) and high yields of P15N > 30%.[39] In this technique, microtesla magnetic fields are employed to promote polarization transfer from p-H2 derived protons to heteronuclei. SABRE-SHEATH was shown to hyperpolarize metronidazole,[40] the FDA-approved antibiotic that can be safely administered in large doses of up to several grams.[41, 42] This antibiotic is structurally similar to many nitroimidazole-based 18F-labeled radiotracers, employed for hypoxia imaging in cancer and other diseases.[43–45] More recently, we have demonstrated that spin-relays[46, 47] can provide efficient polarization of 15N-15N spin-spin coupled sites in metronidazole with −15NO2 group gaining high levels of polarization (>16%) and very long T1 of ~10 min.[48] The −NO2 group undergoes reduction during hypoxia sensing process, but 18F-labeled radiotracers are not capable of distinguishing metabolized compounds, and radiotracer’s clearance from surrounding tissues takes several hours.[49] As a result, clearance time of ~2 h is typically required prior to nitroimidazole-based positron emission tomography (PET) scan to gain sufficient contrast with hypoxic region. 15N-HP nitroimidazoles can potentially mitigate this shortcoming, and enable fast hypoxia scan (<5 min) without ionizing radiation.
Figure 1.

Schematics of spin-relayed[46] SABRE-SHEATH[37, 38] 15N hyperpolarization of [15N3]nimorazole. Both p-H2 and [15N3]nimorazole are in chemical exchange with the Ir complex. In the microtesla magnetic field, hyperpolarization is transferred from p-H2-derived hydride protons to the 15N nuclei of coordinated substrate via spin-relayed mechanism. Dissociation of HP [15N3]nimorazole from the Ir complex allows to obtain free HP [15N3]nimorazole in solution.
Here, we report on synthesis and SABRE-SHEATH hyperpolarization study of [15N3]nimorazole. Although it has been originally developed as an antibiotic against anaerobic infections,[41] it is currently under a Phase 3 clinical study in Europe as a radio-sensitizing agent for the treatment of head and neck cancer.[50, 51] If successful, nimorazole will be integrated in the first-line of cancer chemotherapy treatment. We envision that HP [15N3]nimorazole can be simultaneously employed for both tumor radiosensitization and also as an imaging agent to report on tumor hypoxia status.
Results and Discussion
Typical SABRE-SHEATH experiments employed ~600 μL of solution containing ~100 mM of substrate and ~5 mM of SABRE pre-catalyst [Ir(IMes)(COD)]Cl[23, 53] in CD3OD, placed in a medium-wall 5 mm NMR tube tightly connected with 1/4 inch outer-diameter Teflon tube. NMR tube was placed into the flask filled with water to maintain the constant temperature, and the flask itself was placed in the MuMETAL magnetic shield with the ability to control magnetic field inside via an additional solenoid magnet. The SABRE catalyst was activated via bubbling of p-H2 gas through the solution at 20 standard cubic centimeters per minute (sccm) flow rate and 7.8 bar pressure for ~2 h. Typical SABRE-SHEATH experiments were performed using the following procedure. p-H2 gas was bubbled through the solution at 70 sccm flow rate and 7.8 bar pressure for 60 s. Then gas flow through the solution was terminated by opening the valve on the by-pass line (Scheme 2). NMR tube was pulled out from the magnetic shield, wiped with a paper towel and placed into the probe of 1.4 T Nanalysis NMReady-60PRO bench-top NMR spectrometer for 15N NMR detection.
Scheme 2.

Scheme of experimental setup.
Our initial 15N SABRE-SHEATH experiments using nimorazole with natural abundance of 15N nuclei were unsuccessful, and 1H detection of SABRE-HP samples yielded very small 1H signal enhancements of ca. 3–4-fold (Figure S1). Therefore, we performed the synthesis of isotopically labeled [15N3]nimorazole to increase 15N enrichment from ~0.4% (natural abundance) to ~99% (Scheme 1). In the first step, [15N2]imidazole was prepared with 50–60% yield via condensation of glyoxal, formaldehyde and 15NH4Cl used as a source of 15N isotope enrichment.[52] Subsequent nitration with H15NO3/H2SO4 allowed to obtain 4(5)-[15N]nitro-[1,3-15N2]imidazole with 30% yield. Next, this compound was alkylated with 4-(2-chloroethyl)morpholine in the presence of K2CO3. Due to tautomerism of the imidazole heterocycle, both nitrogen atoms possess nucleophilic properties, leading to unselective alkylation. Thus, the yield of [15N3]nimorazole was only 20% while the yield of its isomer (here termed [15N3]isonimorazole for convenience) was 60% in the last step. The prevalence of isonimorazole is not unexpected since the corresponding nitrogen atom is more nucleophilic than the other one because it is further away from electron accepting nitro group. Moreover, steric hindrance effects favor the formation of isonimorazole compared to nimorazole. The two 15N-labeled isomers were separated chromatographically. See SI for additional synthetic information and spectral characterization of the products.
Scheme 1.

Synthesis of [15N3]nimorazole (4-(2-(5-[15N]nitro-1H-[1,3-15N2]imidazol-1-yl)ethyl)morpholine) and side product [15N3]isonimorazole (4-(2-(4-[15N]nitro-1H-[1,3-15N2]imidazol-1-yl)ethyl)morpholine). Blue and red colors denote the sources of 15N isotope label.
SABRE-SHEATH experiments using [15N3]nimorazole demonstrated efficient 15N hyperpolarization on all three 15N nuclei—the representative 15N NMR spectrum is shown in Figure 2a. The finding that all three 15N sites have nearly equal %P15N (ranging from 1.9% to 3.2%) is similar to the recent study of spin-relayed[46, 47] (via the network formed by 2J15N-15N) polarization transfer in [15N3]metronidazole.[48] Although the maximum %P15N level for [15N3]nimorazole was lower than that for [15N3]metronidazole by a factor of ~5.3 (3.2% vs. 17%[48]), the concentration of to-be-hyperpolarized substrate (and catalyst) was ~4.2-fold greater in [15N3]nimorazole studies. Moreover, a somewhat higher hyperpolarization efficiency of [15N3]metronidazole may also be tentatively attributed to the electron donating properties of methyl group (lacking in [15N3]nimorazole heterocycle) likely modulating the efficiency of SABRE-SHEATH polarization transfer process. We conclude that 15N-15N spin-relayed polarization transfer is a relatively general phenomenon, which is now documented in the molecular motifs of [15N3]nimorazole, [15N3]metronidazole and [15N3]isonimorazole (see below). In this spin-relayed mechanism,[48] 15N-3 nuclear spin is hyperpolarized first via polarization transfer from p-H2-derived hydrides (Figure 1). Second, polarization is transferred from 15N-3 to 15N-1 via 2J15N-15N. Finally, polarization is transferred from 15N-1 to 15NO2 via another 2J15N-15N (Figure 1).[48]
Figure 2.

(a) Single-scan 15N NMR spectrum of HP [15N3]nimorazole. Conditions: ~140 mM substrate, ~7 mM catalyst, 52 °C, 7.8 bar p-H2 pressure, 70 sccm gas flow rate, 0.4 μT magnetic field inside the magnetic shield. (b) 15N NMR spectrum of thermally polarized neat [15N]pyridine acquired with 16 signal accumulations. Note this spectrum intensity is divided by the number of signal accumulations and scaled by a factor of 128. The spectra were acquired on a 1.4 T bench-top NMR spectrometer. The 15N NMR spectrum of HP [15N3]nimorazole recorded on a 7.05 T NMR spectrometer is presented in Figure S2.
To maximize the polarization levels, >2 h of activation of SABRE complex[38, 55] in the presence of [15N3]nimorazole is required (Figure 3a). This activation time is comparable to that observed with [15N3]metronidazole under similar conditions.[56] The magnetic field profile of polarization transfer in microtesla field range exhibits a clear maximum at ~0.4 μT for all three 15N sites (Figure 3b). The maximum position is the same as that for [15N3]metronidazole,[48] further supporting the similarity in J coupling network between the polarizing 15N spins in these two molecules. Moreover, the microtesla magnetic field profile is similar for all three 15N nuclei of [15N3]nimorazole, serving as an additional proof of spin-relayed polarization transfer mechanism discussed above. Furthermore, the dynamics of 15N polarization buildup showed that characteristic exponential buildup times are similar (Tb = 23–29 s) for the three 15N nuclei (Figure 3c) at optimal BT of ~0.4 μT. Note when sub-optimal BT is employed (e.g., <0.04 μT as in Figure S3a), the equilibrium %P15N is achieved faster (Tb = 3.1–3.5 s) albeit yielding much lower %P15N by at least several folds.
Figure 3.

SABRE-SHEATH hyperpolarization of [15N3]nimorazole at ~5 mM catalyst and ~100 mM substrate initial concentrations: (a) SABRE-SHEATH catalyst activation kinetics in the presence of [15N3]nimorazole at 20 °C, 7.8 bar p-H2 pressure, 20 sccm gas low rate and 0.4 μT. (b) SABRE-SHEATH magnetic field profile of [15N3]nimorazole obtained at 20 °C, 7.8 bar p-H2 pressure and 70 sccm p-H2 flow rate. (c) SABRE-SHEATH polarization buildup of [15N3]nimorazole obtained at 20 °C, 7.8 bar p-H2 pressure, 70 sccm gas low rate and 0.4 μT. (d) Relaxation kinetics of HP [15N3]nimorazole at 0.4 μT. (e) Relaxation kinetics of HP [15N3]nimorazole at ~10 μT. (f) Relaxation kinetics of HP [15N3]nimorazole at 1.4 T. (g) The gas flow rate dependence of SABRE-SHEATH polarization of [15N3]nimorazole nitro group obtained at different pressures, 20 °C and 0.4 μT. The inset shows pressure dependence of nitro group 15N polarization obtained at 99 sccm gas flow rate. (h) Temperature dependence of SABRE-SHEATH polarization of [15N3]nimorazole obtained at 7.8 bar p-H2 pressure, 70 sccm gas low rate and 0.4 μT.
Next, the 15N T1 relaxation dynamics of HP [15N3]nimorazole was studied at three relevant static magnetic fields. At ~0.4 μT (the optimal polarization transfer field for [15N3]nimorazole), the signals of three 15N nuclei decayed with approximately the same characteristic relaxation time T1 = 29–33 s (Figure 3d). This 15N T1 clustering at ~0.4 μT (and similar type of clustering at ~10 μT in Figure 3e, and < 0.04 μT in Figure S3b) is important for two reasons. First, it is consistent with the fact that these three 15N nuclei are strongly coupled at microtesla magnetic fields and thus continuously share magnetization with each other via 15N-15N spin relays.[47, 48] Second and more importantly, our recent study of [15N3] and [15N2]metronidazole isotopologues clearly demonstrated the detrimental role of 14N quadrupolar nucleus presence in the spin-relayed network.[56] In this case, two-bond 15N-14N spin-spin coupling results in a significantly lower effective 15N T1 for all (remaining) 15N sites in [15N2]metronidazole versus [15N3]metronidazole: 15N T1 clustering of 8–9 seconds versus 23–31 seconds respectively under similar (1:20) catalyst-to-substrate ratio as the one studied here.[56] [15N3]nimorazole T1 relaxation cluster of 29–33 seconds (Figure 3d) overlaps well with [15N3]metronidazole cluster[56] rather than with [15N2]metronidazole. We rationalize this very important observation in a way that while two-bond 15N-14N spin-spin coupling results in efficient 14N-rendered quadrupolar depolarization effects, the corresponding three-bond coupling has a negligible effect. This is advantageous in the context of development of 15N HP contrast agents with SABRE-SHEATH approach as remote N-sites don’t require 15N isotopic labeling to maximize polarization efficiency of the 15N spins.
Once SABRE-SHEATH polarization transfer was completed, the samples were transferred from the microtesla field to the 1.4 T bench-top NMR spectrometer through the Earth’s magnetic field. In this experimental protocol, relaxation of 15N nuclei of [15N3]nimorazole at the Earth’s field (ca. 10 μT in our lab, apparently due to presence of various magnetic equipment etc.) may have an impact on the resultant 15N polarization levels detected at 1.4 T due to T1 relaxation process during the sample transfer (Figure 3e). At the clinically relevant field of 1.4 T of bench-top NMR spectrometer, 15N T1 values for the three 15N sites were significantly lengthened to 51 ± 4 s for 15N-3, 109 ± 4 s for 15N-1 and 353 ± 13 s for 15NO2 sites, respectively. Although slower relaxation for the 15N nuclei of 15NO2 is expected due to relative isolation of this 15N site from any protons, this long T1 of nearly 6 min is remarkable as it allows storing 15N polarization for tens of minutes (Figure 3f).
Parahydrogen flow rate dependence of 15N SABRE-SHEATH hyperpolarization of [15N3]nimorazole demonstrated that polarization plateaus after ~50 sccm gas flow rate (Figure 3g). Increase in p-H2 pressure leads to a nearly linear %P15N increase (Figure 3g) indicating that %P15N can be further enhanced at elevated p-H2 partial pressure.[55] The temperature dependence of [15N3]nimorazole polarization exhibits a clear maximum at ~54 °C – with %P15N being an order of magnitude greater than that at ~20 °C. Thus, temperature optimization is crucial for efficient SABRE-SHEATH hyperpolarization of [15N3]nimorazole. This observation is different from the optimal temperature of nearly all other SABRE-SHEATH substrates studied to date with optimal %P15N achieved at room temperature of ~20 °C[25, 27, 38, 40] with the exception of [15N2]imidazole[57] (tOPT > 75 °C)—this is likely explained by the different kinetics of nimorazole dissociation from Ir complex.
Similar measurements were carried out at lower concentrations (1 mM catalyst and 20 mM [15N3]nimorazole, Figure S4), and this different concentration regime rendered overall similar results. The addition of co-substrate (pyridine[58]) to the initial solution promoted the rate of SABRE complex activation (Figure S5). However, the maximum in the magnetic field profile of SABRE-SHEATH polarization shifted from ~0.4 μT to ~0.15–0.20 μT, indicating that the active complex likely contained both pyridine and [15N3]nimorazole molecules. Characteristic polarization buildup and relaxation times at microtesla magnetic field were similar to those observed without a co-substrate (compare Figure 3 and Figure S5). On the other hand, 15N relaxation at the Earth’s magnetic field was more efficient, while relaxation at 1.4 T was slower in the presence of pyridine (Figure 3 and Figure S5). Taken altogether, addition of pyridine as a co-substrate led to significant %P15N decrease by 2–6-fold.
We have also explored the possibility to hyperpolarize the isomer of [15N3]nimorazole ([15N3]isonimorazole) using SABRE-SHEATH. In this compound, the nitro group is located closer to the donor N atom that can coordinate to Ir complex than in nimorazole and metronidazole. This proximity of nitro group is expected to complicate coordination of the substrate to the metal center due to both steric and electronic effects.[54, 59] Nevertheless, we have demonstrated that SABRE-SHEATH hyperpolarization of all 15N nuclei of [15N3]isonimorazole is possible, resulting in up to 0.47% polarization (Figure S6). This observation paves the way to SABRE-SHEATH hyperpolarization of other biologically active nitroimidazole derivatives containing nitro group in the α position to the nitrogen atom with a lone electron pair. This range of compounds includes prospective antitumor prodrug evofosfamide (a.k.a. TH-302[60]), antibiotics azomycin and benznidazole, immunosuppressive drug azathioprine and many others.
Dependence of SABRE-SHEATH polarization efficiency of [15N3]isonimorazole on magnetic field, polarization buildup time and temperature was investigated, as well as relaxation of 15N hyperpolarization at three different magnetic fields. Magnetic field profile demonstrated the maximum at 0.25 μT (Figure S8a), which is lower than corresponding maximum for [15N3]nimorazole. Characteristic polarization buildup time for [15N3]isonimorazole (Figure S8b) was also lower than for [15N3]nimorazole (~12–13 s vs. 23–29 s). Temperature dependence of SABRE-SHEATH polarization for [15N3]isonimorazole was also dramatically different: the broad maximum was observed at ~30–43 °C (Figure S8f), in contrast to a sharper peak at ~54 °C for [15N3]nimorazole. Comparing the efficiency of hyperpolarization for the three 15N sites, 15NO2 group exhibited the highest polarization levels, similar to [15N3]nimorazole. However, in case of [15N3]isonimorazole two 15N nuclei in imidazole ring exhibited similar polarization levels, while for [15N3]nimorazole 15N-1 site was polarized more efficiently.
Relaxation of 15N nuclei in [15N3]isonimorazole at microtesla and the Earth’s magnetic field was faster than in [15N3]nimorazole. Similar values (19–21 s and ~18–24 s) were observed at both magnetic fields under study (Figures S8c and S8d). On the other hand, at 1.4 T relaxation of [15N3]isonimorazole 15N nuclei was slower compared to [15N3]nimorazole (Figure S8e): 90 ± 10 s for 15N-3 site, 128 ± 14 s for 15N-1 site and 417 ± 27 s for 15NO2 group, respectively. The long relaxation time for 15NO2 group of [15N3]isonimorazole signifies that other nitroimidazole compounds yet to be hyperpolarized by SABRE will likely exhibit a similarly long relaxation times of HP 15N sites.
The feasibility of 2D sub-second 15N MRI visualization of HP [15N3]nimorazole was demonstrated. Figure 4 presents 2D 15N TrueFISP MR images of a 5 mm NMR tube containing ~110 mM HP [15N3]nimorazole acquired on a 9.4 T MRI scanner. Note the very high spatial (0.5×0.5 mm2/pixel) and temporal (356 ms) resolution that can be potentially achievable with HP 15N MRI of [15N3]nimorazole – first two images in each series are shown for two different projection views. This proof-of-principle demonstration opens the prospects for utilization of SABRE 15N hyperpolarization of [15N3]nimorazole in bioimaging applications such as reporting on tumor hypoxia status in a manner similar to that of nitroimidazole-based PET tracers.[43–45]
Figure 4.

15N TrueFISP MRI of HP [15N3]nimorazole (~110 mM) in a 5 mm NMR tube using multi-nuclear 9.4 T MRI scanner. Frequency offset was adjusted to the signal of 15NO2 group. The imaging parameters: spectral width (SW) = 5 kHz, repetition time (TR) = 14 ms, echo time (TE) = 7 ms, flip angle = 30˚. (a–b) Axial projection (the first and the second scan): scan acquisition time = 356 ms, matrix size = 32×32 (zero-filled to 512×512), FOV = 1.6 cm × 1.6 cm, spatial resolution = 0.5×0.5 mm2/pixel. (c–d) Coronal projection (the first and the second scan): scan acquisition time = 684 ms, matrix size = 64×64 (zero-filled to 512×512), FOV = 7.8 cm × 7.8 cm, spatial resolution = 1.2×1.2 mm2/pixel. Full sets of 16 consecutively recorded MR images are presented in Figure S9 (axial projection) and Figure S10 (coronal projection). 15N NMR spectra acquired at the similar conditions demonstrated P15N of 0.78% for the 15NO2 group and P15N of 0.33% for the 15N-3 site (Figure S11).
Biomedical Outlook
The reported herein study was performed in CD3OD because it is readily available and it has been employed by the vast majority of published in vitro studies. For future in vivo applications, we envision hyperpolarization in a more biocompatible ethanol solvent (SABRE hyperpolarization of nicotinamide in ethanol has been well documented[55]) followed by rapid catalyst removal using functionalized resins[39, 61] and reconstitution to aqueous medium via rapid vacuum distillation of ethanol down to less that 10% content. These two steps should take less than 0.5 minute based on our experience resulting in relatively small polarization losses for all three 15N sites. The presence of 10% ethanol is a generally accepted clinical practice in the context of PET tracers’ administration with up to 20 mL dose.[62] The development of bio-injectable catalyst-free formulation is the next key step for SABRE-SHEATH hyperpolarization technique.
Although catalyst activation takes ~2 h, it can be performed well in advance of the hyperpolarization procedure with normal hydrogen (as the materials are stable for many hours), and the bolus of [15N3]nimorazole can be hyperpolarized in ~1 min. Although the optimum polarization temperature is ~50–60 °C (Figure 3h), this is not a translational challenge—our previous experience with other parahydrogen-hyperpolarized 13C compounds produced at 68 °C shows that the HP liquid equilibrates to below 37 °C at the time of the actual in vivo injection.[63, 64] We note that the produced dose of HP [15N3]nimorazole in NMR tube of ~50 μmoles is certainly sufficient for future pilot in vivo studies in small rodents—previous successful studies with HP 13C injectable agents employed ~5 μmole dose in mice.[9, 64] We envision that the injected nimorazole will undergo cellular uptake within one minute based on the pharmacodynamics data for structurally similar nitroimidazole compounds.[45] The subsequent metabolism of nitroimidazole compounds in hypoxic cells leads to formation of nitroso and hydroxylamine intermediates and amino-derivative metabolic products, Figure 5.[65] The ab initio calculations of 15N chemical shifts using Gaussian’09 (see SI for details) for nimorazole and its metabolic reduction products reveal that the associated 15N chemical shifts of all labelled 15N sites are very sensitive to reduction process. While the nitro-group nitrogen is the most sensitive one (with nearly 800 ppm dynamic range), it may not necessarily be the most useful sensing site for hydroxylamino- and amino-derivatives because of the anticipated decrease in 15N T1 due to directly attached protons. On the other hand, while the dynamic range of the other two sites is less dramatic (12 ppm and 20 ppm respectively), these two 15N sites will likely retain their long T1 in vivo. To summarize, the relaxation properties and the 15N chemical shift dispersion of metabolic products bode well for future in vivo studies of using HP [15N3]nimorazole as an injectable hypoxia sensor.
Figure 5.

Metabolism of nitroimidazole compounds on the example of nimorazole in hypoxic environment.[65] Color-coded values of 15N chemical shifts (in ppm) were computed for aqueous media using Gaussian’09 ab initio calculations (see SI for details).
Conclusion
To summarize, we report on efficient synthesis and SABRE-SHEATH hyperpolarization of 15N-labeled [15N3]nimorazole. The HP state with %P15N of up to 3.2% 15N polarization is created on all three 15N sites, and it persists for tens of minutes with 15N T1 decay of 5.9 ± 0.2 min at clinically relevant field of 1.4 T. This report paves the way to future theragnostic imaging applications of [15N3]nimorazole and other HP nitroimidazoles as cancer chemotherapeutic agents and as tumor hypoxia sensors that may significantly aid to advance personalized care of cancer patients.
The relaxation dynamics results presented here provide valuable insights for the rational development of schemes and molecular substrates for efficient spin-relayed 15N SABRE-SHEATH hyperpolarization. Although synthesis yields a side product, we were able to successfully separate and purify both compounds. The feasibility of SABRE-SHEATH 15N hyperpolarization of the synthesized isomer of [15N3]nimorazole with 15NO2 group in α position to donor nitrogen atom of imidazole ring opens opportunities for efficient hyperpolarization of a wide range of other biomedically interesting molecules with very long-lived HP state retained by NO2 group. For example, HP azomycin can potentially act as a suitable pH sensor (pKa ~ 7.0),[57] and TH-302[66] can be employed as a theragnostic agent for cytotoxic chemotherapeutics selectively activating on hypoxic tumors, which are typically difficult to treat using conventional chemotherapies.
Supplementary Material
Acknowledgements
O.G.S. thanks the Russian Foundation for Basic Research (Grant 19-33-60045) for the support of mechanistic studies with the use of parahydrogen. E.Y.C. thanks the following awards for funding support: NSF CHE-1904780 and NIH 1R21CA220137. N.V.C., A.S., K.V.K. and I.V.K. thank the Russian Foundation for Basic Research (grants 17-54-33037, 18-33-20019 and 19-53-12013) for financial support and the Russian Ministry of Science and Higher Education (Project AAAA-A16-116121510087-5). We dedicate this work to the memory of Dr. Kirill V. Kovtunov, our mentor, colleague and friend.
Footnotes
Supporting information for this article is given via a link at the end of the document.
Institute and/or researcher Twitter usernames: https://twitter.com/mmt_lab, https://twitter.com/waynestatechem, https://twitter.com/OlegSalnikovNMR
References
- [1].Nikolaou P, Goodson BM, Chekmenev EY, Chem. Eur. J 2015, 21, 3156–3166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Kovtunov KV, Pokochueva EV, Salnikov OG, Cousin S, Kurzbach D, Vuichoud B, Jannin S, Chekmenev EY, Goodson BM, Barskiy DA, Koptyug IV, Chem. Asian J 2018, 13, 1857–1871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Golman K, in’t Zandt R, Thaning M, Proc. Natl. Acad. Sci. U. S. A 2006, 103, 11270–11275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Kurhanewicz J, Vigneron DB, Brindle K, Chekmenev EY, Comment A, Cunningham CH, DeBerardinis RJ, Green GG, Leach MO, Rajan SS, Rizi RR, Ross BD, Warren WS, Malloy CR, Neoplasia 2011, 13, 81–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Brindle KM, J. Am. Chem. Soc 2015, 137, 6418–6427. [DOI] [PubMed] [Google Scholar]
- [6].Day SE, Kettunen MI, Gallagher FA, Hu DE, Lerche M, Wolber J, Golman K, Ardenkjaer-Larsen JH, Brindle KM, Nat. Med 2007, 13, 1382–1387. [DOI] [PubMed] [Google Scholar]
- [7].Albers MJ, Bok R, Chen AP, Cunningham CH, Zierhut ML, Zhang VY, Kohler SJ, Tropp J, Hurd RE, Yen Y-F, Nelson SJ, Vigneron DB, Kurhanewicz J, Cancer Res. 2008, 68, 8607–8615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Nelson SJ, Kurhanewicz J, Vigneron DB, Larson PEZ, Harzstark AL, Ferrone M, van Criekinge M, Chang JW, Bok R, Park I, Reed G, Carvajal L, Small EJ, Munster P, Weinberg VK, Ardenkjaer-Larsen JH, Chen AP, Hurd RE, Odegardstuen LI, Robb FJ, Tropp J, Murray JA, Sci. Transl. Med 2013, 5, 198ra108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Cavallari E, Carrera C, Sorge M, Bonne G, Muchir A, Aime S, Reineri F, Sci. Rep 2018, 8, 8366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Merritt M, Harrison C, Storey C, Jeffrey F, Sherry A, Malloy C, Proc. Natl. Acad. Sci. U. S. A 2007, 104, 19773–19777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Comment A, J. Magn. Reson 2016, 264, 39–48. [DOI] [PubMed] [Google Scholar]
- [12].Kurhanewicz J, Vigneron DB, Ardenkjaer-Larsen JH, Bankson JA, Brindle K, Cunningham CH, Gallagher FA, Keshari KR, Kjaer A, Laustsen C, Mankoff DA, Merritt ME, Nelson SJ, Pauly JM, Lee P, Ronen S, Tyler DJ, Rajan SS, Spielman DM, Wald L, Zhang X, Malloy CR, Rizi R, Neoplasia 2019, 21, 1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Ardenkjaer-Larsen JH, Fridlund B, Gram A, Hansson G, Hansson L, Lerche MH, Servin R, Thaning M, Golman K, Proc. Natl. Acad. Sci. U. S. A 2003, 100, 10158–10163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Ardenkjaer-Larsen JH, J. Magn. Reson 2016, 264, 3–12. [DOI] [PubMed] [Google Scholar]
- [15].Bornet A, Melzi R, Perez Linde AJ, Hautle P, van den Brandt B, Jannin S, Bodenhausen G, J. Phys. Chem. Lett 2013, 4, 111–114. [DOI] [PubMed] [Google Scholar]
- [16].Vuichoud B, Milani J, Bornet A, Melzi R, Jannin S, Bodenhausen G, J. Phys. Chem. B 2014, 118, 1411–1415. [DOI] [PubMed] [Google Scholar]
- [17].Eichhorn TR, Takado Y, Salameh N, Capozzi A, Cheng T, Hyacinthe J-N, Mishkovsky M, Roussel C, Comment A, Proc. Natl. Acad. Sci. U. S. A 2013, 110, 18064–18069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Milani J, Vuichoud B, Bornet A, Melzi R, Jannin S, Bodenhausen G, Rev. Sci. Instrum 2017, 88, 015109. [DOI] [PubMed] [Google Scholar]
- [19].Hövener J-B, Pravdivtsev AN, Kidd B, Bowers CR, Glöggler S, Kovtunov KV, Plaumann M, Katz-Brull R, Buckenmaier K, Jerschow A, Reineri F, Theis T, Shchepin RV, Wagner S, Bhattacharya P, Zacharias NM, Chekmenev EY, Angew. Chem. Int. Ed 2018, 57, 11140–11162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Adams RW, Duckett SB, Green RA, Williamson DC, Green GGR, J. Chem. Phys 2009, 131, 194505. [DOI] [PubMed] [Google Scholar]
- [21].Atkinson KD, Cowley MJ, Duckett SB, Elliott PIP, Green GGR, López-Serrano J, Khazal IG, Whitwood AC, Inorg. Chem 2009, 48, 663–670. [DOI] [PubMed] [Google Scholar]
- [22].Atkinson KD, Cowley MJ, Elliott PIP, Duckett SB, Green GGR, Lopez-Serrano J, Whitwood AC, J. Am. Chem. Soc 2009, 131, 13362–13368. [DOI] [PubMed] [Google Scholar]
- [23].Cowley MJ, Adams RW, Atkinson KD, Cockett MCR, Duckett SB, Green GGR, Lohman JAB, Kerssebaum R, Kilgour D, Mewis RE, J. Am. Chem. Soc 2011, 133, 6134–6137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Rayner PJ, Duckett SB, Angew. Chem. Int. Ed 2018, 57, 6742–6753. [DOI] [PubMed] [Google Scholar]
- [25].Colell JFP, Logan AWJ, Zhou Z, Shchepin RV, Barskiy DA, Ortiz GX, Wang Q, Malcolmson SJ, Chekmenev EY, Warren WS, Theis T, J. Phys. Chem. C 2017, 121, 6626–6634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Iali W, Roy SS, Tickner BJ, Ahwal F, Kennerley AJ, Duckett SB, Angew. Chem. Int. Ed 2019, 58, 10271–10275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Gemeinhardt ME, Limbach MN, Gebhardt TR, Eriksson CW, Eriksson SL, Lindale JR, Goodson EA, Warren WS, Chekmenev EY, Goodson BM, Angew. Chem. Int. Ed 2020, 59, 418–423. [DOI] [PubMed] [Google Scholar]
- [28].Theis T, Ortiz GX , Logan AWJ, Claytor KE, Feng Y, Huhn WP, Blum V, Malcolmson SJ, Chekmenev EY, Wang Q, Warren WS, Sci. Adv 2016, 2, e1501438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Kovtunov KV, Koptyug IV, Fekete M, Duckett SB, Theis T, Joalland B, Chekmenev EY, Angew. Chem. Int. Ed 2020, 59, 17788–17797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Iali W, Rayner PJ, Duckett SB, Sci. Adv 2018, 4, eaao6250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].McCormick J, Korchak S, Mamone S, Ertas YN, Liu Z, Verlinsky L, Wagner S, Glöggler S, Bouchard L-S, Angew. Chem. Int. Ed 2018, 57, 10692–10696. [DOI] [PubMed] [Google Scholar]
- [32].Bales L, Kovtunov KV, Barskiy DA, Shchepin RV, Coffey AM, Kovtunova LM, Bukhtiyarov AV, Feldman MA, Bukhtiyarov VI, Chekmenev EY, Koptyug IV, Goodson BM, J. Phys. Chem. C 2017, 121, 15304–15309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Nonaka H, Hata R, Doura T, Nishihara T, Kumagai K, Akakabe M, Tsuda M, Ichikawa K, Sando S, Nat. Commun 2013, 4, 2411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Sarkar R, Comment A, Vasos PR, Jannin S, Gruetter R, Bodenhausen G, Hall H, Kirik D, Denisov VP, J. Am. Chem. Soc 2009, 131, 16014–16015. [DOI] [PubMed] [Google Scholar]
- [35].Cudalbu C, Comment A, Kurdzesau F, van Heeswijk RB, Uffmann K, Jannin S, Denisov V, Kirik D, Gruetter R, Phys. Chem. Chem. Phys 2010, 12, 5818–5823. [DOI] [PubMed] [Google Scholar]
- [36].Jiang W, Lumata L, Chen W, Zhang S, Kovacs Z, Sherry AD, Khemtong C, Sci. Rep 2015, 5, 9104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Theis T, Truong ML, Coffey AM, Shchepin RV, Waddell KW, Shi F, Goodson BM, Warren WS, Chekmenev EY, J. Am. Chem. Soc 2015, 137, 1404–1407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Truong ML, Theis T, Coffey AM, Shchepin RV, Waddell KW, Shi F, Goodson BM, Warren WS, Chekmenev EY, J. Phys. Chem. C 2015, 119, 8786–8797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Kidd BE, Gesiorski JL, Gemeinhardt ME, Shchepin RV, Kovtunov KV, Koptyug IV, Chekmenev EY, Goodson BM, J. Phys. Chem. C 2018, 122, 16848–16852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Barskiy DA, Shchepin RV, Coffey AM, Theis T, Warren WS, Goodson BM, Chekmenev EY, J. Am. Chem. Soc 2016, 138, 8080–8083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Roy RB, Laird SM, Heasman L, Br. J. Vener. Dis 1975, 51, 281–284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [42].Erickson SH, Oppenheim GL, Smith GH, Obstet. Gynecol 1981, 57, 48–50. [PubMed] [Google Scholar]
- [43].Hendrickson K, Phillips M, Smith W, Peterson L, Krohn K, Rajendran J, Radiother. Oncol 2011, 101, 369–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [44].Hirata K, Kobayashi K, Tamaki N, Springer Japan, Tokyo, 2016, pp. 229–249. [Google Scholar]
- [45].Komar G, Seppänen M, Eskola O, Lindholm P, Grönroos TJ, Forsback S, Sipilä H, Evans SM, Solin O, Minn H, J. Nucl. Med 2008, 49, 1944–1951. [DOI] [PubMed] [Google Scholar]
- [46].Shchepin RV, Jaigirdar L, Theis T, Warren WS, Goodson BM, Chekmenev EY, J. Phys. Chem. C 2017, 121, 28425–28434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [47].Shchepin RV, Jaigirdar L, Chekmenev EY, J. Phys. Chem. C 2018, 122, 4984–4996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [48].Shchepin RV, Birchall JR, Chukanov NV, Kovtunov KV, Koptyug IV, Theis T, Warren WS, Gelovani JG, Goodson BM, Shokouhi S, Rosen MS, Yen Y-F, Pham W, Chekmenev EY, Chem. Eur. J 2019, 25, 8829–8836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [49].Fleming IN, Manavaki R, Blower PJ, West C, Williams KJ, Harris AL, Domarkas J, Lord S, Baldry C, Gilbert FJ, Br. J. Cancer 2015, 112, 238–250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [50].ClinicalTrials.gov, Clinical trial number NCT01507467: “IAEA-HypoX. Accelerated Radiotherapy With or Without Nimorazole in Squamous Cell Carcinoma of the Head and Neck”. [Google Scholar]
- [51].Henk JM, Bishop K, Shepherd SF, Radiother. Oncol 2003, 66, 65–70. [DOI] [PubMed] [Google Scholar]
- [52].Shchepin RV, Barskiy DA, Coffey AM, Feldman MA, Kovtunova LM, Bukhtiyarov VI, Kovtunov KV, Goodson BM, Koptyug IV, Chekmenev EY, ChemistrySelect 2017, 2, 4478–4483. [Google Scholar]
- [53].Barskiy DA, Kovtunov KV, Koptyug IV, He P, Groome KA, Best QA, Shi F, Goodson BM, Shchepin RV, Coffey AM, Waddell KW, Chekmenev EY, J. Am. Chem. Soc 2014, 136, 3322–3325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [54].Rayner PJ, Norcott P, Appleby KM, Iali W, John RO, Hart SJ, Whitwood AC, Duckett SB, Nat. Commun 2018, 9, 4251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [55].Truong ML, Shi F, He P, Yuan B, Plunkett KN, Coffey AM, Shchepin RV, Barskiy DA, Kovtunov KV, Koptyug IV, Waddell KW, Goodson BM, Chekmenev EY, J. Phys. Chem. B 2014, 18 13882–13889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [56].Birchall JR, Kabir MSH, Salnikov OG, Chukanov NV, Kovtunov KV, Koptyug IV, Gelovani JG, Goodson BM, Pham W, Chekmenev EY, Chem. Comm 2020, 56, 9098–9101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [57].Shchepin RV, Barskiy DA, Coffey AM, Theis T, Shi F, Warren WS, Goodson BM, Chekmenev EY, ACS Sensors 2016, 1, 640–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [58].Shchepin RV, Barskiy DA, Mikhaylov DM, Chekmenev EY, Bioconjugate Chem. 2016, 27, 878–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [59].Shchepin RV, Truong ML, Theis T, Coffey AM, Shi F, Waddell KW, Warren WS, Goodson BM, Chekmenev EY, J. Phys. Chem. Lett 2015, 6, 1961–1967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [60].Liapis V, Zinonos I, Labrinidis A, Hay S, Ponomarev V, Panagopoulos V, Zysk A, DeNichilo M, Ingman W, Atkins GJ, Findlay DM, Zannettino ACW, Evdokiou A, Cancer medicine 2016, 5, 534–545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [61].Barskiy DA, Ke LA, Li X, Stevenson V, Widarman N, Zhang H, Truxal A, Pines A, J. Phys. Chem. Lett 2018, 9, 2721–2724. [DOI] [PubMed] [Google Scholar]
- [62].Serdons K, Verbruggen A, Bormans G, J. Nucl. Med 2008, 49, 2071. [DOI] [PubMed] [Google Scholar]
- [63].Shchepin RV, Coffey AM, Waddell KW, Chekmenev EY, Anal. Chem 2014, 86, 5601–5605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [64].Coffey AM, Shchepin RV, Truong ML, Wilkens K, Pham W, Chekmenev EY, Anal. Chem 2016, 88, 8279–8288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [65].Masaki Y, Shimizu Y, Yoshioka T, Tanaka Y, Nishijima K.-i., Zhao S, Higashino K, Sakamoto S, Numata Y, Yamaguchi Y, Tamaki N, Kuge Y, Sci. Rep 2015, 5, 16802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [66].Stokes AM, Hart CP, Quarles CC, Tomography 2016, 2, 229–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
